Storage news ticker - September 4

Cadence announced that its PHY and controller IP for the PCI Express (PCIe) 6.0 specification, implemented in the TSMC N3 process, achieved first-pass success at the PCI-SIG compliance workshop in late July, marking the industry's first official event for PCIe 6.0 compliance testing with the x8 subsystem tested at 64 GT/s. Cloudera introduced native GPU acceleration for Apache Spark 4.1 in Cloudera Data Engineering, powered by Nvidia CUDA-X cuDF, enabling zero-code acceleration across hybrid environments. Everspin and Teledyne HiRel partnered to bring Everspin's PERSYST STT-MRAM to aerospace and defense programs, while Lenovo unveiled a concept notebook with Frore AirJet solid-state active cooling, and Kioxia outlined plans to use fast NAND as a DRAM alternative via CXL modules, potentially doubling memory capacity and improving performance by 30 percent.

FLASH Storage news ticker - September 4 Cadence announced its PHY and controller IP for the PCI Express PCIe 6.0 specification, implemented in the TSMC N3 process, achieved first-pass success at the recent PCI-SIG compliance workshop held in late July. The workshop marked the industry’s first official event for PCIe 6.0 specification compliance testing, with Cadence’s x8 subsystem solution tested at the full PCIe 6.0 specification speed of 64 GT/s. … Cloudera announced native GPU acceleration for Apache Spark 4.1 in Cloudera Data Engineering, enabled by the Nvidia CUDA-X library, cuDF. The cuDF plug-in for Apache Spark will support the just-announced Cloudera Anywhere Cloud, which is designed to enable organizations to accelerate Spark workloads without rewriting PySpark or SQL code. This helps data teams prepare AI-ready data faster while reducing cloud infrastructure costs across hybrid environments. Cloudera Data Engineering accelerated by CUDA-X libraries will deliver: - Zero-code GPU acceleration for Apache Spark 4.1 workloads - Faster ETL and data preparation for analytics and AI - Lower cloud infrastructure costs through shorter compute runtimes - Built-in deployment with no manual driver configuration - Enterprise security and governance through the Cloudera Unified Data Fabric - Consistent performance across public cloud, private cloud, sovereign cloud, and on-premises environments Unlike GPU acceleration offerings limited to a single cloud provider, Cloudera extends these capabilities across hybrid environments while preserving consistent governance and operations. Organizations can accelerate Spark workloads wherever their data resides— whether that is on premises, in the clouds, or at the edge— without sacrificing security or flexibility. … Everspin and Teledyne HiRel have partnered to bring Everspin's PERSYST STT-MRAM into Teledyne HiRel's memory portfolio for aerospace, defense and space programs. That gives customers a path to evaluate and design in MRAM, with screening, qualification documentation and obsolescence support, for functions that have traditionally relied on NOR flash or battery-backed SRAM. … Lenovo announced the world's first ultra-slim ultra-quiet notebook featuring Frore AirJet Solid-State Active Cooling. Lenovo’s Active Flow Thermal Solution is the world’s first application of AirJet designed into a concept notebook … Infinidat, now a Lenovo company, continues to grow and grow and grow. Its recruiting and has 17 openings in its customer facing teams. More here https://t.co/1AaFKGgFUu . … Kioxia has plans to use fast NAND as a DRAM alternative, according to Japan’s Nikkei https://asia.nikkei.com/business/tech/semiconductors/kioxia-outlines-plan-to-capture-ai-demand-with-its-new-high-speed-memory-chip outlet. It intends to use a CXL Compute Express Link module wth memory-access speed across a PCE link. Combining high-speed NAND with a CXL controller module reduces data-read latency to less than one-tenth that of conventional NAND products. Replacing part of a system's DRAM with CXL modules can double memory capacity and improve performance by 30 percent, according to Kioxia. Kioxia also plans to begin sampling an SSD with enhanced read-and-write performance in 2028. … NetApp and AWS announced that AWS Transform now supports Amazon FSx for NetApp ONTAP. AWS Transform is an agentic AI service that automates the discovery, planning, and migration of workloads to the cloud. Now generally available, customers can use AWS Transform to migrate attached storage from their on-premises and cloud estates directly to Amazon FSx for ONTAP. Amazon FSx for ONTAP supported by AWS Transform gives customers the option to migrate block storage as part of the same migration wave that handles compute and network, eliminating the need for intermediate storage platforms, separate migration tools, and the additional cost and risk they introduce. … Enterprise file services and data management supplier Peer Software announced customer success results from Würth Elektronik's deployment of PeerGFS, enabling the global electronics manufacturer to provide fast, reliable access to engineering files across North America, Europe and Asia while protecting file integrity for mission-critical CAD workflows. PeerGFS synchronises approximately 800 GB of active engineering data in real time across four Windows-based file servers spanning three continents. The platform provides distributed file locking to prevent conflicting edits, centralised management and built-in high availability, allowing project teams to collaborate without delays or data integrity concerns. Prior to deploying PeerGFS, engineers accessing CAD files from a centralised server could wait more than five minutes for files to open. Today, they access local replicas that load in seconds, dramatically improving productivity while ensuring reliable file locking across regions to prevent costly rework caused by overlapping changes. … OceanBase has been named the top distributed transactional database vendor in Asia and Oceania based on 2025 revenue, according to a newly released Omdia report, Distributed Transactional Database in Asia & Oceania - 2026. Capturing a 15.5 percent market share, OceanBase outpaced major cloud service providers, underscoring its leadership in supporting mission-critical workloads across the region. … Open-source vector search and database supplier Qdrant’s Kumar Shivendu, from its core engineering team has published a blog https://www.kshivendu.dev/blog/token-storage and a paper https://arxiv.org/abs/2608.02376 that rethinks text storage and compression for systems where AI agents are the primary users. Basically he found that we can achieve 1.5-2.5x compression out of the box if you use a language model's tokenizer and it gives you orders of magnitudes lower read and write latency from an LLM's perspective because it can't read UTF-8 bytes, it needs token IDs and tokenization is expensive. The compression can go even higher with some tricks he explained in his blog. Agents have already started reading and writing more data than humans so he proposes that we introduce proper ways to read and write token IDs in our databases and filesystems. … Quantum announced the appointment of Hiral A. Patel as its Chief Accounting Officer and principal accounting officer, replacing Laura Nash whois transitioning to an advisor position. Patel most recently served as Chief Accounting Officer at Pep Boys, an automotive service and retail company, where she led external and internal financial reporting, tax compliance and strategy, and the accounting processes supporting the company's recent divestiture. As part of Patel’s compensation package and as an inducement material to her acceptance of employment with the company, the Leadership and Compensation Committee of Quantum’s Board of Directors approved the grant of 50,000 restricted stock units, which will vest in three equal annual installments. … Snowflake announced its FY’27 Q2 earnings results, “reinforcing that the Agentic Enterprise runs on Snowflake.” It delivered: - Product revenue of $1.49 billion, with growth accelerating to 37 percent year-over-year — third straight quarter of acceleration, adding seven points since the start of FY’27 - Q2 non-GAAP operating margin expanded over 400 basis points year-over-year to 15 percent - Raising FY’27 outlook from 31 percent to 36 percent year-over-year, and raising FY’27 non-GAAP operating margin guidance from 13.5 percent to 14.5 percent - Added 692 net-new customers in Q2, up 32 percent year-over-year, including 14 new Global 2000 customers - 828 customers with trailing 12-month product revenue greater than $1 million, and 48 customers crossing this threshold in Q2 … Synology has released ActiveProtect Manager APM 2.0, an update to its ActiveProtect data protection platform. Additions in APM 2.0: - Broader platform support: Nutanix AHV, Proxmox VE, Amazon EC2, Azure VM and Google Workspace, alongside existing VMware, Hyper‑V, physical, SaaS and database workloads - Proactive ransomware defence: AI/ML anomaly detection and pre‑recovery malware scans, with alerts to SIEM/SOAR via syslog, SMTP or SNMP traps - Stronger enterprise controls: Optional storage‑level encryption and expanded role‑based access controls - More remote storage options: Azure Blob Storage and additional Synology NAS models for tiered, long‑term retention APM 2.0 adds proactive cyber resilience capabilities — including AI/ML‑based anomaly detection and pre‑recovery malware scanning — to help organisations restore only clean, trustworthy data. Note: these two features are scheduled to arrive later AI/ML‑based anomaly detection and pre‑recovery malware scanning are scheduled to arrive later. … Parallel file system supplier VDURA and New Mexico State University NMSU announced that NMSU's new VDURA deployment is complete and in full production. The system, live as of August 2026, provides the university's research community with a high-performance, highly durable data platform purpose-built for AI and high-performance computing workloads. NMSU, which achieved Carnegie R1 status in 2025, placing it among the top research universities in the United States, supports more than $141 million in annual research expenditures across fields including aerospace, agriculture, cybersecurity, water science and biomedical research. The deployment combines NVMe flash and high-density HDD capacity in VDURA's mixed-fleet architecture, delivering flash-level responsiveness for active research data alongside economical bulk capacity, all in a single namespace on the university's InfiniBand fabric. With VDURA's software-defined architecture and subscription model, NMSU can scale capacity and performance as research demands grow. …